Background: We conducted a meta-analysis to determine a practical observation time for detecting a biphasic reaction after resolution of the initial anaphylactic reaction.
Methods: A systematic literature search identified studies on adult patients with anaphylaxis and a subsequent biphasic reaction due to various causes that contained sufficient data to extract outcomes. The outcomes were pooled using a random-effects model.
Results: Twelve studies with a total of 2,890 adult patients with anaphylaxis and 143 patients with a biphasic reaction were included. In terms of the pooled negative predictive value, 1 h of observation achieved a 95.0% negative predictive value and ≥6 h of observation provided a 97.3% negative predictive value (95% CI: 95.0-98.5). The negative predictive value for a biphasic reaction increased with a longer observation time after initial anaphylaxis, and the increasing trend slowed down from 6 h of observation time. The pooled additional incidence rates of biphasic reactions per 100 person-hours after 1- and 4-h observations were 0.45 (95% CI: 0.20-1.04) and 0.41 (95% CI: 0.19-0.87), respectively. After > 8-12 h of postanaphylactic observation, the negative predictive value reached > 98%, while the additional incidence per 100 person-hours was < 0.10.
Conclusions: An observation time of ≥6 h after resolution of an initial anaphylaxis symptom can exclude recurrence of a secondary reaction in > 95% of patients. Although longer observation periods resulted in the detection of more biphasic reactions, 6-12 h of observation time would be practical, supporting current relevant guidelines.
Keywords: Anaphylaxis; Biphasic reaction; Meta-analysis; Observation.
© 2019 S. Karger AG, Basel.